Abductive Reasoning with Uncertainty∗

Abductive reasoning in general describes the process of discovering hypotheses and rules that would entail a given conclusion. Abductive reasoning consists of assessing the likelihood that a specific hypothesis entails a given conclusion. Abductive reasoning based on probabilities is used in many disciplines, such as medical diagnostics, where medical test results combined with conditional probabilities are used to determine the likelihood of possible diseases. In this paper we focus on abductive reasoning in subjective logic. The advantage of our approach over a purely probabilistic approach is that degrees of ignorance can be explicitly included as input and during the analysis.